You’re no doubt familiar with the minimum wage debate. Advocates of an increased minimum wage on the left argue that a higher wage floor is essentially a free lunch that will raise low-end household incomes and reduce turnover. Opponents warn that setting the wage floor too high will reduce net job growth and price a nontrivial number of less-skilled workers, including young workers, out of the formal labor market. What is clear is that calls for a higher minimum wage have been a godsend for a left that finds itself intellectually exhausted. Though the public is wary of expanding government, large majorities favor a higher minimum wage as a seemingly straightforward solution to wage stagnation. And liberals have been quite happy to play along with the notion that a higher minimum wage is the cure-all that many Americans believe it to be.
But have you heard about the maximum wage debate? In Vox, Matt Yglesias makes the case for a maximum wage. More precisely, he is making the case for an effective maximum wage, or tax rates that are so high they deter firms from offering high salaries in the first place. Though I don’t find Yglesias’s case very convincing, I suspect we’ll be hearing more arguments like it from liberal thinkers and activists in the years to come. So for that reason alone, Yglesias’s case deserves a closer look.
Rather than call for a hard cap, Yglesias is really calling for tax rates so high as to be confiscatory, and he identifies the ultra-high top marginal tax rates of the years immediately following the Second World War as a model.
“During the 90 percent top income tax rate,” Yglesias writes, “for a firm to put an extra $100 in the pocket of a top executive required them to pay another $1,000 in salary.” And so firms chose to instead “give modest raises to five separate middle managers.” This strategy did not raise much revenue, but it did distribute compensation more evenly within firms. For Yglesias, revenue is not the goal of a “super-tax.” Instead, his goal is to change America’s political economy by, among other things, limiting the influx of talented workers into ultra-high-wage professions, checking the rent-seeking that (allegedly) has driven the increase in executive compensation, and encouraging firms to divvy up their compensation expenditures in a more egalitarian fashion. And as evidence in favor of this approach, Yglesias observes that “the pre-Reagan trend of productivity growth in the American economy was faster than the post-Reagan trend.”
First, Yglesias’s comparison of pre-Reagan productivity growth and post-Reagan productivity growth neglects the possibility that in the absence of tax reform, productivity growth might have been lower still. In his essay on “Frontier Economics,” Brink Lindsey argues that the key difference between the period from 1947 to 1973, when the average annual increase in productivity in the U.S. was 2.9 percent, and the period from 1980 to 2006, when it was only 2 percent, is that the immediate postwar decades offered myriad opportunities for what he calls “imitative growth,” in which existing technologies and business models are deployed across the economy, particularly in underdeveloped regions in the South and West. This is much like the catch-up growth we see in less-affluent societies, where tried-and-true strategies for raising productivity are deployed until the society in question reach middle-income status, at which point growth tends to plateau. The latter period saw “the exhaustion of relatively easy opportunities for imitative growth in the United States and other advanced economies,” and so continued growth has depended on business model innovation, an area in which the U.S. has, in relative terms, at least, excelled. Comparing the growth record of the post-Reagan U.S. to that of the postwar golden age is not unlike comparing the growth record of an advanced market economy with a poor country in the throes of catch-up growth. Even the Soviet bloc economies saw substantial economic growth in the immediate postwar decades as these societies urbanized and as workers shifted from the agricultural to the industrial sector. That, alas, is a trick that can only be pulled off once.
Now let’s consider America’s postwar tax regime in more detail. Last spring, Arpit Gupta revisited the high tax rates of the 1950s. He observed that very few people paid the high-end official rates, a fact that could be attributed to the fact that firms were deterred from offering high compensation, as in Yglesias’s model. Yet Gupta also found that average income-tax rates remained stable from the 1950s to the 2000s. The highest earners of the 1950s, whose marginal dollars were subject to the highest rates, paid average tax rates comparable to those of the highest earners. When we consider the incentives facing the high earners of this era, it is important to take into account the loopholes and deductions that allowed them to shield their income from tax.
It is also true, however, that the tax code of the postwar years imposed much higher corporate and estate taxes. Everyone agrees that the burden of the estate tax was borne primarily by the wealthy. But there is an ongoing debate over how to understand the burden of corporate taxes. Some analysts maintain that the burden of corporate taxes falls entirely on shareholders while others believe that the burden falls at least in part on the employees of firms, for whom corporate taxes translate into lower wages. Yglesias accepts that capital income will have to be shielded from his super-tax, to ensure that the reallocation of capital proceeds apace. So it is worth noting that, according to Thomas Piketty and Emmanuel Saez, at least, the main reason the tax code of the postwar years was so progressive is that it featured much higher taxes on capital income. This isn’t a problem for Yglesias’s argument, as he is convinced that the high rates of the postwar years succeeded in preventing firms from paying high salaries.
It’s not clear to me that Yglesias is thinking about the rise in executive compensation in the right way. While it is true that executive compensation was lower in this era, it is also true that the market capitalization of the largest firms was also much smaller. One possibility is that as the market capitalization of big companies has increased, the stakes associated with finding and retaining an effective CEO have risen. This in turn has contributed to an intensified competition for the best managers, or rather those perceived to be the best managers, which has led to an increase in executive compensation. There are many questions we can’t reliably answer. For example, it is possible that the market capitalization would never have increased as much as it did in the counterfactual world in which tax rates remained at their postwar highs. Or it is possible that loopholes and deductions would have shielded high earners from taxes in this counterfactual world, and the arms race for managerial talent would have played out much as it did in our own world.
Yglesias rejects this benign interpretation of executive compensation growth, noting that “it is now a commonplace of progressive discourse to argue that executive compensation growth largely reflects rent-seeking that could be rolled back without impairing managerial talent.” The paper he cites, by Joshua Bivens and Lawrence Mishel of the Economic Policy Institute, is interesting, and it merits a (forthcoming) discussion of its own. For now, I’ll just note that Jim Manzi scrutinized Piketty’s closely related argument that executive compensation growth, and rising inequality more broadly, can be attributed in to the bargaining power of top executives, and he found it highly implausible.
Going even further, Yglesias cites a recent argument from Benjamin Lockwood, Charles Nathanson, and Glen Weyl that “by increasing the financial incentive for top talent to pursue careers in finance and law rather than teaching and research, the Reagan tax reforms reduced overall economic output while increasing the pre-tax share of income earned by top earners.” Lockwood, Nathanson, and Weyl base their claim on a mechanistic model of the economy. In the absence of the Reagan tax cuts, they posit that many of the workers who went into finance and management would have instead taken up positions in academia, engineering, and teaching; and their model finds that this shift away from research-oriented professions reduced social welfare by 1 to 2 percent. Note that they base this calculation on a generous assessment of the positive externalities associated with research. To their credit, Lockwood, Nathanson, and Weyl note that some of their results strike them as “implausibly large,” and they acknowledge that their reading of the literature is “inevitably partial.” They anticipate that future research will find that “the existing literature overestimates externalities and thus our magnitudes,” and I agree with them. When Yglesias writes “rather than giving the middle class a smaller slice of a bigger pie and making everyone better off, these reforms gave the rich a larger slice of a smaller pie and made only them better off,” he doesn’t make it clear that Lockwood, Nathanson, and Weyl are tentatively suggesting that the reforms in question might have made the pie smaller by 1 to 2 percentage points under assumptions that even they allow strain credulity.
Like Yglesias, I believe that there are serious problems with corporate governance in America. Where we part company is over his apparent conviction that confiscatory taxes are an important part of the solution. I won’t belabor the case against ultra-high marginal tax rates — Arpit Gupta has summarized some of the scholarly research on the impact of high rates on long-run growth. Rather, I’ll just suggest that we focus more directly on how the tax code encourages excessive leverage and how our regulatory regime shields incumbents from competition from the new firms that give rise to new business models. Those are causes that, hopefully, we can all agree on.